A switch device detection method, device, equipment and readable storage medium

Abstract

The present application discloses a switch device detection method, device, equipment and readable storage medium. The present application obtains the state information of the switch device, performs dimension reduction processing on the state information by principal component analysis, obtains the eigenvector matrix, calculates the average value and linear regression trend coefficient of the eigenvector matrix, determines the preliminary remaining service life according to the average value and the linear regression trend coefficient, inputs the state information and the preliminary remaining service life into a life calculation model, and obtains the final remaining service life of the switch device output by the life calculation model. The present application scheme achieves the purpose of detecting the remaining service life of the switch device by inputting the state information of the switch device and the corresponding preliminary remaining service life into the life calculation model, and obtains the final remaining service life of the switch device output by the life calculation model.

Description

Technical Field

[0001] The present application relates to the field of distribution network detection, and in particular to a switchgear detection method, apparatus, device and readable storage medium. Background Art

[0002] With the development of social science and technology, the distribution network operation system has been gradually improved. As a key equipment in the distribution network operation system, the switch has the dual responsibilities of "control" and "protection". The operating reliability of the switch is directly related to the reliable operation of the power system, so its daily maintenance is very important.

[0003] In the daily maintenance of the switch, the opening and closing coil circuit in the switch operating accessories is the main control circuit for the opening and closing actions of the switch. Its normal operation is the key guarantee for the overall operation of the switch. However, the main detection method for its daily operation and maintenance is still subjective detection. By manually observing the appearance of the opening and closing coil and its interlocking mechanical structure, whether the circuit is broken, etc., the faults that can be identified by the naked eye are still difficult to detect for the hidden mechanical problems contained inside. Due to the large installed capacity of switches, complex working environment, variable working temperature, and serious aging of equipment, these factors will cause the switch to fail during operation. So far, how to detect the remaining service life of the switch equipment has become a concern. Summary of the invention

[0004] In view of this, the present application provides a switch device detection method, apparatus, device and readable storage medium.

[0005] In order to achieve the above objectives, the proposed solution is as follows:

[0006] A switch device detection method, comprising:

[0007] Get status information of switchgear;

[0008] Performing dimensionality reduction processing on the state information using principal component analysis to obtain a eigenvector matrix;

[0009] Calculating the average value and linear regression trend coefficient of the eigenvector matrix;

[0010] Determine preliminary remaining useful life based on the mean value and linear regression trend coefficient;

[0011] Inputting the state information and the preliminary remaining service life into a service life calculation model to obtain a final remaining service life of the switch device output by the service life calculation model;

[0012] The life calculation model is trained by using sample status information and a preliminary remaining service life of the sample determined based on the sample status information as training samples, and using the final remaining service life of the switch device as a training label.

[0013] Optionally, the state information is subjected to dimensionality reduction processing using a principal component analysis method to obtain a feature vector matrix, including:

[0014] Converting the state information into two-dimensional data information;

[0015] Calculating the two-dimensional data information to obtain a covariance matrix;

[0016] Calculating the covariance matrix using principal component analysis to obtain the eigenvectors and eigenvalues ​​of the two-dimensional data information;

[0017] The multiple eigenvectors corresponding to the largest eigenvalue are combined into an eigenvector matrix.

[0018] Optionally, the calculating the average value and the linear regression trend coefficient of the eigenvector matrix includes:

[0019] The average value and linear regression trend coefficient of the eigenvector matrix are calculated using the following formula:

[0020]

[0021]

[0022] Among them, a represents the average value of the eigenvector matrix, m represents the number of eigenvectors, b represents the eigenvector, i represents the dimension of the eigenvector, L represents the preset window width parameter, and s represents the linear regression trend coefficient.

[0023] Optionally, determining a preliminary remaining useful life based on the average value and the linear regression trend coefficient includes:

[0024] According to the average value and the linear regression trend coefficient, the corresponding preliminary remaining useful life is determined from a preset window decomposition table, wherein the window decomposition table pre-stores the corresponding relationship between the average value and the linear regression trend coefficient and the preliminary remaining useful life.

[0025] Optionally, the preset process of the window decomposition table includes:

[0026] Obtaining maintenance information of the switchgear;

[0027] Selecting a new switchgear having less maintenance information than the switchgear as a sample;

[0028] Obtaining the sample status information of the sample from the current moment until the sample ages and loses performance, and the corresponding duration, wherein the corresponding duration is the initial remaining service life of the sample at the current moment;

[0029] Converting the sample state information into a feature vector matrix;

[0030] Converting the eigenvector matrix and the corresponding duration into a two-dimensional function image;

[0031] The two-dimensional function image is intercepted by sliding a window based on a preset width parameter to obtain a target two-dimensional function image;

[0032] Calculating the average value and linear regression trend coefficient of the eigenvector matrix contained in the target two-dimensional function image;

[0033] The average value and linear regression trend coefficient of the eigenvector matrix contained in the target two-dimensional function image and the preliminary remaining useful life of the target two-dimensional function image corresponding to the time length are recorded in an initial window decomposition table.

[0034] Optionally, the method further includes:

[0035] The final remaining service life is displayed.

[0036] A switchgear detection device, comprising:

[0037] A status information acquisition unit, used to acquire status information of the switch device;

[0038] An information dimension reduction unit, used for performing dimension reduction processing on the state information by using a principal component analysis method to obtain a eigenvector matrix;

[0039] A feature calculation unit, used to calculate the average value and linear regression trend coefficient of the feature vector matrix;

[0040] A preliminary life calculation unit, used for determining a preliminary remaining useful life according to the average value and the linear regression trend coefficient;

[0041] a final life calculation unit, configured to input the state information and the preliminary remaining life into a life calculation model, and obtain a final remaining life of the switch device output by the life calculation model;

[0042] The life calculation model is trained by using sample status information and a preliminary remaining service life of the sample determined based on the sample status information as training samples, and using the final remaining service life of the switch device as a training label.

[0043] Optionally, the information dimension reduction unit includes:

[0044] A first information dimension reduction subunit, used for converting the state information into two-dimensional data information;

[0045] A second information dimension reduction subunit is used to calculate the two-dimensional data information to obtain a covariance matrix;

[0046] A third information dimension reduction subunit is used to calculate the covariance matrix by using a principal component analysis method to obtain the eigenvector and eigenvalue of the two-dimensional data information;

[0047] The fourth information dimension reduction subunit is used to form an eigenvector matrix from multiple eigenvectors corresponding to the largest eigenvalue.

[0048] A data encryption and decryption device, characterized in that it comprises a memory and a processor;

[0049] The memory is used to store programs;

[0050] The processor is used to execute the program to implement each step of the aforementioned switch device detection method.

[0051] A readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, each step of the aforementioned switch device detection method is implemented.

[0052] It can be seen from the above technical scheme that the embodiments of the present application provide a switch device detection method, device, equipment and readable storage medium. The present application scheme inputs the status information of the switch device and the corresponding preliminary remaining service life into the life calculation model to obtain the final remaining service life of the switch device output by the life calculation model, thereby achieving the purpose of detecting the remaining service life of the switch device. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A flow chart of a switch device detection method provided in an embodiment of the present application;

[0054] Figure 2 A flow chart of a state information dimensionality reduction method provided in an embodiment of the present application;

[0055] Figure 3 A flowchart of a method for presetting a window decomposition table provided in an embodiment of the present application;

[0056] Figure 4 A schematic diagram of a window decomposition two-dimensional function image scene provided in an embodiment of the present application;

[0057] Figure 5 A schematic diagram of the structure of a switch device detection device provided in an embodiment of the present application;

[0058] Figure 6A hardware structure block diagram of a switch device detection device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0060] Figure 1 A flow chart of a switch device detection method provided in an embodiment of the present application, the method may include the following steps:

[0061] Step S100: Acquire status information of a switch device.

[0062] Specifically, the switching device can be a primary switching device of a distribution network, and the status information of the switching device can be obtained through the distribution network detection system. The status information can include dynamic parameter information and static parameter information, such as the moment when the iron core of the switching device starts to move, the moment when the iron core hits the lock, the moment when the current rises to a steady-state current, the moment when the auxiliary contact is disconnected, the moment when the current drops to 0, the coil current when the iron core starts to move, the coil current when the iron core hits the lock, the coil steady-state current, rated current, rated temperature rise, rated short-time withstand current, rated peak withstand current, rated short-circuit duration, etc.

[0063] Step S110: Utilize principal component analysis to perform dimension reduction processing on the state information to obtain a eigenvector matrix.

[0064] Specifically, considering that there are many dynamic parameter information and static parameter information in the state information, it inevitably contains a lot of redundant information, which may affect the accuracy of subsequent calculation results. Therefore, the principal component analysis method can be used to reduce the dimension of the state information, and the obtained eigenvector matrix can be used for subsequent calculation steps.

[0065] Step S120, calculating the average value and linear regression trend coefficient of the eigenvector matrix.

[0066] Specifically, the average value and linear regression trend coefficient of the eigenvector matrix may be used to represent the change trend of the eigenvector in the state information, and the average value and linear regression trend coefficient of the eigenvector matrix may be calculated.

[0067] Step S130: determining a preliminary remaining useful life according to the average value and the linear regression trend coefficient.

[0068] Specifically, the preliminary remaining service life may be a preliminary life test result of the switch device, and the preliminary remaining service life of the switch device may be determined based on an average value and a linear regression trend coefficient.

[0069] Step S140: input the state information and the preliminary remaining service life into a service life calculation model to obtain a final remaining service life of the switch device output by the service life calculation model.

[0070] Specifically, the life calculation model can be trained using sample state information and the preliminary remaining service life of the samples determined based on the sample state information as training samples, and using the final remaining service life of the switching device as training labels. The state information and the preliminary remaining service life of the switching device can be input into the life calculation model to obtain the final remaining service life. The life calculation model can adopt a neural network model. The structure of the neural network model adopted can be multiple, such as convolutional neural network (CNN), back propagation network (BP), recurrent attention neural network (RA-CNN) and CenterNet network, etc., or a network structure after improving the existing network.

[0071] In an embodiment of the present application, by inputting the status information of the switching device and the corresponding preliminary remaining service life into the life calculation model, the final remaining service life of the switching device output by the life calculation model is obtained, thereby achieving the purpose of detecting the remaining service life of the switching device.

[0072] In some embodiments of the present application, Figure 2 As shown, the process of step S110, using the principal component analysis method to reduce the dimension of the state information to obtain the characteristic vector matrix, is introduced below. The process may include:

[0073] Step S111: convert the state information into two-dimensional data information.

[0074] Specifically, in order to standardize various data information in the state information for convenience of calculation, the state information may be converted into two-dimensional data information, for example, the value of the state information may be converted into a range of [-1, 1].

[0075] Step S112: Calculate the two-dimensional data information to obtain a covariance matrix.

[0076] Specifically, the relative change trend of the variables of the two-dimensional data information can be observed through the covariance matrix, and the covariance matrix can be obtained by calculating the two-dimensional data information.

[0077] Step S113: Calculate the covariance matrix using principal component analysis to obtain the eigenvectors and eigenvalues ​​of the two-dimensional data information.

[0078] Specifically, the screening characteristics of the principal component analysis method can be used to extract eigenvectors and their corresponding eigenvalues ​​from the covariance matrix, and the principal component analysis method can be used to calculate the covariance matrix to obtain the eigenvectors and eigenvalues ​​of the two-dimensional data information.

[0079] Step S114: compose a plurality of eigenvectors corresponding to the largest eigenvalue into an eigenvector matrix.

[0080] Specifically, the eigenvalues ​​may be sorted in a preset order of magnitude, and a plurality of eigenvectors corresponding to the largest eigenvalues ​​may be selected to form an eigenvector matrix.

[0081] The embodiment of the present application completes the dimensionality reduction processing of the state information through the above steps to obtain a feature vector matrix.

[0082] Further, in some embodiments of the present application, the process of calculating the average value and the linear regression trend coefficient of the eigenvector matrix in step S120 is introduced, and the process may include the following steps:

[0083] The average value and linear regression trend coefficient of the eigenvector matrix are calculated using the following formula:

[0084]

[0085]

[0086] Specifically, a represents the average value of the eigenvector matrix, m represents the number of eigenvectors, b represents the eigenvector, i represents the dimension of the eigenvector, L represents the preset window width parameter, and s represents the linear regression trend coefficient.

[0087] In some embodiments of the present application, the process of determining the preliminary remaining useful life according to the average value and the linear regression trend coefficient in step S130 is introduced, and the process may include the following steps:

[0088] Step S131 : determining the corresponding preliminary remaining useful life from a preset window decomposition table according to the average value and the linear regression trend coefficient.

[0089] Specifically, the window decomposition table pre-stores the correspondence between the average value and the linear regression trend coefficient and the preliminary remaining useful life, and the corresponding preliminary remaining useful life can be determined from the window decomposition table according to the average value and the linear regression trend coefficient.

[0090] Further, in some embodiments of the present application, the preset process of the window decomposition table is introduced below. Figure 3 As shown, the process may include the following steps:

[0091] Step S200: Obtain maintenance information of the switch device.

[0092] Specifically, the maintenance information may include information on inspection, maintenance, and repair of the switchgear according to a set time period.

[0093] Step S210: Select a new switch device whose maintenance information is less than that of the switch device as a sample.

[0094] Specifically, since the time points of installing the switchgear are different, the corresponding maintenance information will also be different. In order to select switchgear with a remaining service life longer than the switchgear to be tested as a sample, a new switchgear with less maintenance information than the switchgear can be selected as a sample.

[0095] Step S220: Obtain the sample status information from the current moment of the sample until the sample ages and loses performance, and the corresponding time duration.

[0096] Specifically, the corresponding duration may be the preliminary remaining service life of the sample at the current moment, and the preliminary remaining service life of the corresponding duration may be obtained by acquiring the sample state information from the current moment until the sample ages and loses performance.

[0097] Step S230: Convert the sample state information into a feature vector matrix.

[0098] Specifically, step S230 corresponds to step S110 in the aforementioned embodiment, and details thereof are referred to the aforementioned introduction, which will not be repeated here.

[0099] Step S240: converting the feature vector matrix and the corresponding duration into a two-dimensional function image.

[0100] Specifically, in order to reflect the corresponding relationship between the eigenvector matrix and the corresponding time length, the eigenvector matrix and the corresponding time length can be converted into a two-dimensional function image, for example Figure 4 As shown, the horizontal axis represents the corresponding duration, the vertical axis represents the eigenvector in the eigenvector matrix, and the duration unit can be set as required, such as hours.

[0101] Step S250: intercepting the two-dimensional function image by sliding a window based on a preset width parameter to obtain a target two-dimensional function image.

[0102] Specifically, the range of the feature vectors to be calculated can be selected by setting the window width parameter, and the target two-dimensional function image can be obtained by sliding the window to intercept the two-dimensional function image. For example, Figure 4 The window width parameter in can be set to L, and the width parameter 20 can be set.

[0103] Step S260, calculating the average value and linear regression trend coefficient of the eigenvector matrix contained in the target two-dimensional function image.

[0104] Specifically, step S260 corresponds to step S120 in the aforementioned embodiment, and details are given in the aforementioned introduction, which will not be repeated here.

[0105] Step S270, recording the average value and linear regression trend coefficient of the eigenvector matrix contained in the target two-dimensional function image, and the preliminary remaining useful life of the target two-dimensional function image corresponding to the time length into an initial window decomposition table.

[0106] Specifically, the preliminary remaining service life of the target two-dimensional function image corresponding to the time length can be obtained by subtracting the horizontal coordinate of the lower right corner of the window from the preliminary remaining service life of the sample state information to obtain the preliminary remaining service life of the target two-dimensional function image corresponding to the time length. The average value and linear regression trend coefficient of the eigenvector matrix contained in the target two-dimensional function image and the preliminary remaining service life of the target two-dimensional function image corresponding to the time length can be recorded in the initial window decomposition table.

[0107] In some embodiments of the present application, the solution may further include the following steps:

[0108] Step S150: displaying the final remaining service life.

[0109] Specifically, the final remaining service life of the switch device can be displayed through a terminal, and the terminal can be a mobile phone, tablet, computer, etc. with a display function.

[0110] The switch device detection device provided in an embodiment of the present application is described below. The switch device detection device described below and the switch device detection method described above can be referenced to each other.

[0111] like Figure 5 As shown, it discloses a structural schematic diagram of a switch device detection device, which may include:

[0112] An information acquisition unit 11, used to acquire status information of a switch device;

[0113] An information dimension reduction unit 12, used to perform dimension reduction processing on the state information using a principal component analysis method to obtain a feature vector matrix;

[0114] A vector calculation unit 13, used to calculate the average value and linear regression trend coefficient of the characteristic vector matrix;

[0115] A preliminary determination unit 14, used to determine a preliminary remaining useful life according to the average value and the linear regression trend coefficient;

[0116] A life calculation unit 15, configured to input the state information and the preliminary remaining service life into a life calculation model, and obtain a final remaining service life of the switch device output by the life calculation model;

[0117] The life calculation model is trained by using sample status information and a preliminary remaining service life of the sample determined based on the sample status information as training samples, and using the final remaining service life of the switch device as a training label.

[0118] Optionally, the information dimension reduction unit 12 may include:

[0119] A first information dimension reduction subunit, used for converting the state information into two-dimensional data information;

[0120] A second information dimension reduction subunit is used to calculate the two-dimensional data information to obtain a covariance matrix;

[0121] A third information dimension reduction subunit is used to calculate the covariance matrix by using a principal component analysis method to obtain the eigenvector and eigenvalue of the two-dimensional data information;

[0122] The fourth information dimension reduction subunit is used to form an eigenvector matrix from multiple eigenvectors corresponding to the largest eigenvalue.

[0123] Optionally, the vector calculation unit 13 may include:

[0124] The first vector calculation subunit is used to calculate the average value and linear regression trend coefficient of the eigenvector matrix using the following formula:

[0125]

[0126]

[0127] Among them, a represents the average value of the eigenvector matrix, m represents the number of eigenvectors, b represents the eigenvector, i represents the dimension of the eigenvector, L represents the preset window width parameter, and s represents the linear regression trend coefficient.

[0128] Optionally, the preliminary determination unit 14 may include:

[0129] The first preliminary determination subunit is used to determine the corresponding preliminary remaining useful life from a preset window decomposition table according to the average value and the linear regression trend coefficient. The window decomposition table pre-stores the corresponding relationship between the average value and the linear regression trend coefficient and the preliminary remaining useful life.

[0130] Optionally, the switch device detection device further includes:

[0131] A table preset module, used for presetting the window decomposition table;

[0132] The preset process of the window decomposition table may include:

[0133] Obtaining maintenance information of the switchgear;

[0134] Selecting a new switchgear having less maintenance information than the switchgear as a sample;

[0135] Obtaining the sample status information of the sample from the current moment until the sample ages and loses performance, and the corresponding duration, wherein the corresponding duration is the initial remaining service life of the sample at the current moment;

[0136] Converting the sample state information into a feature vector matrix;

[0137] Converting the eigenvector matrix and the corresponding duration into a two-dimensional function image;

[0138] The two-dimensional function image is intercepted by sliding a window based on a preset width parameter to obtain a target two-dimensional function image;

[0139] Calculating the average value and linear regression trend coefficient of the eigenvector matrix contained in the target two-dimensional function image;

[0140] The average value and linear regression trend coefficient of the eigenvector matrix contained in the target two-dimensional function image and the preliminary remaining useful life of the target two-dimensional function image corresponding to the time length are recorded in an initial window decomposition table.

[0141] Optionally, the switch device detection device may further include:

[0142] A display unit is used to display the final remaining service life.

[0143] The switch device detection apparatus provided in the embodiment of the present application can be applied to a switch device detection device. The switch device detection device can be a terminal. Figure 6 The hardware structure diagram of the switchgear detection device is shown in FIG. Figure 6, the hardware structure of the switchgear detection device may include: at least one processor 1, at least one communication interface 2, at least one memory 3 and at least one communication bus 4;

[0144] In the embodiment of the present application, the number of the processor 1, the communication interface 2, the memory 3, and the communication bus 4 is at least one, and the processor 1, the communication interface 2, and the memory 3 communicate with each other through the communication bus 4;

[0145] The processor 1 may be a central processing unit CPU, or an application-specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present invention, etc.;

[0146] The memory 3 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory;

[0147] The memory stores a program, and the processor can call the program stored in the memory, wherein the program is used to:

[0148] Get status information of switchgear;

[0149] Performing dimensionality reduction processing on the state information using principal component analysis to obtain a eigenvector matrix;

[0150] Calculating the average value and linear regression trend coefficient of the eigenvector matrix;

[0151] Determine preliminary remaining useful life based on the mean value and linear regression trend coefficient;

[0152] Inputting the state information and the preliminary remaining service life into a service life calculation model to obtain a final remaining service life of the switch device output by the service life calculation model;

[0153] The life calculation model is trained by using sample status information and a preliminary remaining service life of the sample determined based on the sample status information as training samples, and using the final remaining service life of the switch device as a training label.

[0154] Optionally, the detailed functions and extended functions of the program may refer to the above description.

[0155] The embodiment of the present application further provides a storage medium, which may store a program suitable for execution by a processor, wherein the program is used to:

[0156] Get status information of switchgear;

[0157] Performing dimensionality reduction processing on the state information using principal component analysis to obtain a eigenvector matrix;

[0158] Calculating the average value and linear regression trend coefficient of the eigenvector matrix;

[0159] Determine preliminary remaining useful life based on the mean value and linear regression trend coefficient;

[0160] Inputting the state information and the preliminary remaining service life into a service life calculation model to obtain a final remaining service life of the switch device output by the service life calculation model;

[0161] The life calculation model is trained by using sample status information and a preliminary remaining service life of the sample determined based on the sample status information as training samples, and using the final remaining service life of the switch device as a training label.

[0162] Optionally, the detailed functions and extended functions of the program may refer to the above description.

[0163] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0164] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined with each other, and the same or similar parts can be referenced to each other.

[0165] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A switch device detection method, characterized in that: include: Acquiring status information of the switch device, wherein the status information includes dynamic parameter information and static parameter information; The state information specifically includes the moment when the core of the switch starts to move, the moment when the core hits the lock, the moment when the current rises to a steady state, the moment when the auxiliary contact is disconnected, the moment when the current drops to 0, the coil current when the core starts to move, the coil current when the core hits the lock, the coil steady-state current, the rated current, the rated temperature rise, the rated short-time withstand current, the rated peak withstand current, and the rated short-circuit duration; Performing dimensionality reduction processing on the state information using principal component analysis to obtain a eigenvector matrix; Calculating the average value and linear regression trend coefficient of the eigenvector matrix; Determine preliminary remaining useful life based on the mean value and linear regression trend coefficient; Inputting the state information and the preliminary remaining service life into a service life calculation model to obtain a final remaining service life of the switch device output by the service life calculation model; The life calculation model is trained by using the sample state information and the preliminary remaining service life of the sample determined based on the sample state information as training samples, and using the final remaining service life of the switch device as a training label; The calculating the average value and linear regression trend coefficient of the characteristic vector matrix includes: The average value and linear regression trend coefficient of the eigenvector matrix are calculated using the following formula: ； ； in, represents the mean value of the eigenvector matrix, represents the number of eigenvectors, represents the feature vector, represents the dimension of the feature vector, L represents the preset window width parameter, represents the linear regression trend coefficient; Determining the preliminary remaining useful life according to the average value and the linear regression trend coefficient includes: According to the average value and the linear regression trend coefficient, the corresponding preliminary remaining useful life is determined from a preset window decomposition table, wherein the window decomposition table pre-stores the corresponding relationship between the average value and the linear regression trend coefficient and the preliminary remaining useful life; The preset process of the window decomposition table includes: Obtaining maintenance information of the switchgear; Selecting a new switchgear having less maintenance information than the switchgear as a sample; Obtaining the sample status information of the sample from the current moment until the sample ages and loses performance, and the corresponding duration, wherein the corresponding duration is the initial remaining service life of the sample at the current moment; Converting the sample state information into a feature vector matrix; Converting the eigenvector matrix and the corresponding duration into a two-dimensional function image; The two-dimensional function image is intercepted by sliding a window based on a preset width parameter to obtain a target two-dimensional function image; Calculating the average value and linear regression trend coefficient of the eigenvector matrix contained in the target two-dimensional function image; The average value and linear regression trend coefficient of the eigenvector matrix contained in the target two-dimensional function image and the preliminary remaining useful life of the target two-dimensional function image corresponding to the time length are recorded in an initial window decomposition table.

2. The method according to claim 1, characterized in that The state information is subjected to dimensionality reduction processing using the principal component analysis method to obtain a feature vector matrix, including: Converting the state information into two-dimensional data information; Calculating the two-dimensional data information to obtain a covariance matrix; Calculating the covariance matrix using principal component analysis to obtain the eigenvectors and eigenvalues ​​of the two-dimensional data information; The multiple eigenvectors corresponding to the largest eigenvalue are combined into an eigenvector matrix.

3. The method according to claim 1, characterized in that Also includes: The final remaining service life is displayed.

4. A switchgear detection device, characterized in that: include: A state information acquisition unit, used to acquire state information of the switch device, wherein the state information includes dynamic parameter information and static parameter information; The state information specifically includes the moment when the core of the switch starts to move, the moment when the core hits the lock, the moment when the current rises to a steady state, the moment when the auxiliary contact is disconnected, the moment when the current drops to 0, the coil current when the core starts to move, the coil current when the core hits the lock, the coil steady-state current, the rated current, the rated temperature rise, the rated short-time withstand current, the rated peak withstand current, and the rated short-circuit duration; An information dimension reduction unit, used for performing dimension reduction processing on the state information by using a principal component analysis method to obtain a eigenvector matrix; A feature calculation unit, used to calculate the average value and linear regression trend coefficient of the feature vector matrix; A preliminary life calculation unit, used to determine a preliminary remaining useful life according to the average value and the linear regression trend coefficient; a final life calculation unit, configured to input the state information and the preliminary remaining life into a life calculation model, and obtain a final remaining life of the switch device output by the life calculation model; The life calculation model is trained by using the sample state information and the preliminary remaining service life of the sample determined based on the sample state information as training samples, and using the final remaining service life of the switch device as a training label; The calculating the average value and linear regression trend coefficient of the characteristic vector matrix includes: The average value and linear regression trend coefficient of the eigenvector matrix are calculated using the following formula: ； ； in, represents the mean value of the eigenvector matrix, represents the number of eigenvectors, represents the feature vector, represents the dimension of the feature vector, L represents the preset window width parameter, represents the linear regression trend coefficient; Determining the preliminary remaining useful life according to the average value and the linear regression trend coefficient includes: According to the average value and the linear regression trend coefficient, the corresponding preliminary remaining useful life is determined from a preset window decomposition table, wherein the window decomposition table pre-stores the corresponding relationship between the average value and the linear regression trend coefficient and the preliminary remaining useful life; The preset process of the window decomposition table includes: Obtaining maintenance information of the switchgear; Selecting a new switchgear having less maintenance information than the switchgear as a sample; Obtaining the sample status information of the sample from the current moment until the sample ages and loses performance, and the corresponding duration, wherein the corresponding duration is the initial remaining service life of the sample at the current moment; Converting the sample state information into a feature vector matrix; Converting the eigenvector matrix and the corresponding duration into a two-dimensional function image; The two-dimensional function image is intercepted by sliding a window based on a preset width parameter to obtain a target two-dimensional function image; Calculating the average value and linear regression trend coefficient of the eigenvector matrix contained in the target two-dimensional function image; The average value and linear regression trend coefficient of the eigenvector matrix contained in the target two-dimensional function image and the preliminary remaining useful life of the target two-dimensional function image corresponding to the time length are recorded in an initial window decomposition table.

5. The device according to claim 4, characterized in that The information dimension reduction unit comprises: A first information dimension reduction subunit, used for converting the state information into two-dimensional data information; A second information dimension reduction subunit is used to calculate the two-dimensional data information to obtain a covariance matrix; A third information dimension reduction subunit is used to calculate the covariance matrix by using a principal component analysis method to obtain the eigenvector and eigenvalue of the two-dimensional data information; The fourth information dimension reduction subunit is used to form an eigenvector matrix from multiple eigenvectors corresponding to the largest eigenvalue.

6. A data encryption and decryption device, characterized in that: including memory and processor; The memory is used to store programs; The processor is used to execute the program to implement each step of the switch device detection method according to any one of claims 1 to 3.

7. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, each step of the switch device detection method according to any one of claims 1 to 3 is implemented.

Images